Automatic identification of functional clusters in fMRI data using spatial dependence

Sai Ma, Nicolle M. Correa, Xi Lin Li, Tom Eichele, Vince Daniel Calhoun, Tlay Adali

Research output: Contribution to journalArticle

Abstract

In independent component analysis (ICA) of functional magnetic resonance imaging (fMRI) data, extracting a large number of maximally independent components provides a detailed functional segmentation of brain. However, such high-order segmentation does not establish the relationships among different brain networks, and also studying and classifying components can be challenging. In this study, we present a multidimensional ICA (MICA) scheme to achieve automatic component clustering. In our MICA framework, stable components are hierarchically grouped into clusters based on higher order statistical dependencemutual informationamong spatial components, instead of the typically used temporal correlation among time courses. The final cluster membership is determined using a statistical hypothesis testing method. Since ICA decomposition takes into account the modulation of the spatial maps, i.e., temporal information, our ICA-based approach incorporates both spatial and temporal information effectively. Our experimental results from both simulated and real fMRI datasets show that the use of spatial dependence leads to physiologically meaningful connectivity structure of brain networks, which is consistently identified across various ICA model orders and algorithms. In addition, we observe that components related to artifacts, including cerebrospinal fluid, arteries, and large draining veins, are grouped together and encouragingly distinguished from other components of interest.

Original languageEnglish (US)
Article number6009176
Pages (from-to)3406-3417
Number of pages12
JournalIEEE Transactions on Biomedical Engineering
Volume58
Issue number12 PART 1
DOIs
StatePublished - Dec 2011
Externally publishedYes

Fingerprint

Independent component analysis
Brain
Cerebrospinal fluid
Magnetic Resonance Imaging
Modulation
Decomposition
Testing

Keywords

  • Functional magnetic resonance imaging (fMRI)
  • independent component analysis (ICA)
  • multidimensional independent component analysis (MICA)
  • spatial dependence

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Automatic identification of functional clusters in fMRI data using spatial dependence. / Ma, Sai; Correa, Nicolle M.; Li, Xi Lin; Eichele, Tom; Calhoun, Vince Daniel; Adali, Tlay.

In: IEEE Transactions on Biomedical Engineering, Vol. 58, No. 12 PART 1, 6009176, 12.2011, p. 3406-3417.

Research output: Contribution to journalArticle

Ma, Sai ; Correa, Nicolle M. ; Li, Xi Lin ; Eichele, Tom ; Calhoun, Vince Daniel ; Adali, Tlay. / Automatic identification of functional clusters in fMRI data using spatial dependence. In: IEEE Transactions on Biomedical Engineering. 2011 ; Vol. 58, No. 12 PART 1. pp. 3406-3417.
@article{79b2f2f6737b40fda5c91c5a3e1c59c4,
title = "Automatic identification of functional clusters in fMRI data using spatial dependence",
abstract = "In independent component analysis (ICA) of functional magnetic resonance imaging (fMRI) data, extracting a large number of maximally independent components provides a detailed functional segmentation of brain. However, such high-order segmentation does not establish the relationships among different brain networks, and also studying and classifying components can be challenging. In this study, we present a multidimensional ICA (MICA) scheme to achieve automatic component clustering. In our MICA framework, stable components are hierarchically grouped into clusters based on higher order statistical dependencemutual informationamong spatial components, instead of the typically used temporal correlation among time courses. The final cluster membership is determined using a statistical hypothesis testing method. Since ICA decomposition takes into account the modulation of the spatial maps, i.e., temporal information, our ICA-based approach incorporates both spatial and temporal information effectively. Our experimental results from both simulated and real fMRI datasets show that the use of spatial dependence leads to physiologically meaningful connectivity structure of brain networks, which is consistently identified across various ICA model orders and algorithms. In addition, we observe that components related to artifacts, including cerebrospinal fluid, arteries, and large draining veins, are grouped together and encouragingly distinguished from other components of interest.",
keywords = "Functional magnetic resonance imaging (fMRI), independent component analysis (ICA), multidimensional independent component analysis (MICA), spatial dependence",
author = "Sai Ma and Correa, {Nicolle M.} and Li, {Xi Lin} and Tom Eichele and Calhoun, {Vince Daniel} and Tlay Adali",
year = "2011",
month = "12",
doi = "10.1109/TBME.2011.2167149",
language = "English (US)",
volume = "58",
pages = "3406--3417",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",
number = "12 PART 1",

}

TY - JOUR

T1 - Automatic identification of functional clusters in fMRI data using spatial dependence

AU - Ma, Sai

AU - Correa, Nicolle M.

AU - Li, Xi Lin

AU - Eichele, Tom

AU - Calhoun, Vince Daniel

AU - Adali, Tlay

PY - 2011/12

Y1 - 2011/12

N2 - In independent component analysis (ICA) of functional magnetic resonance imaging (fMRI) data, extracting a large number of maximally independent components provides a detailed functional segmentation of brain. However, such high-order segmentation does not establish the relationships among different brain networks, and also studying and classifying components can be challenging. In this study, we present a multidimensional ICA (MICA) scheme to achieve automatic component clustering. In our MICA framework, stable components are hierarchically grouped into clusters based on higher order statistical dependencemutual informationamong spatial components, instead of the typically used temporal correlation among time courses. The final cluster membership is determined using a statistical hypothesis testing method. Since ICA decomposition takes into account the modulation of the spatial maps, i.e., temporal information, our ICA-based approach incorporates both spatial and temporal information effectively. Our experimental results from both simulated and real fMRI datasets show that the use of spatial dependence leads to physiologically meaningful connectivity structure of brain networks, which is consistently identified across various ICA model orders and algorithms. In addition, we observe that components related to artifacts, including cerebrospinal fluid, arteries, and large draining veins, are grouped together and encouragingly distinguished from other components of interest.

AB - In independent component analysis (ICA) of functional magnetic resonance imaging (fMRI) data, extracting a large number of maximally independent components provides a detailed functional segmentation of brain. However, such high-order segmentation does not establish the relationships among different brain networks, and also studying and classifying components can be challenging. In this study, we present a multidimensional ICA (MICA) scheme to achieve automatic component clustering. In our MICA framework, stable components are hierarchically grouped into clusters based on higher order statistical dependencemutual informationamong spatial components, instead of the typically used temporal correlation among time courses. The final cluster membership is determined using a statistical hypothesis testing method. Since ICA decomposition takes into account the modulation of the spatial maps, i.e., temporal information, our ICA-based approach incorporates both spatial and temporal information effectively. Our experimental results from both simulated and real fMRI datasets show that the use of spatial dependence leads to physiologically meaningful connectivity structure of brain networks, which is consistently identified across various ICA model orders and algorithms. In addition, we observe that components related to artifacts, including cerebrospinal fluid, arteries, and large draining veins, are grouped together and encouragingly distinguished from other components of interest.

KW - Functional magnetic resonance imaging (fMRI)

KW - independent component analysis (ICA)

KW - multidimensional independent component analysis (MICA)

KW - spatial dependence

UR - http://www.scopus.com/inward/record.url?scp=82155191306&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=82155191306&partnerID=8YFLogxK

U2 - 10.1109/TBME.2011.2167149

DO - 10.1109/TBME.2011.2167149

M3 - Article

C2 - 21900068

AN - SCOPUS:82155191306

VL - 58

SP - 3406

EP - 3417

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 12 PART 1

M1 - 6009176

ER -